traffico-ticino 0.1.1

A Python package to analyze traffic and air pollution in Ticino

traffico_ticino – Air Pollution and Traffic Analysis in Southern Switzerland

traffico_ticino is a Python package designed to analyze the relationship between vehicular traffic and air pollution in the Canton of Ticino, Switzerland.
It combines observed data from environmental monitoring stations with meteorological inputs and provides forecasting tools for future pollution scenarios under different traffic evolution assumptions.

---

Key Features

• Regression analysis linking traffic intensity to pollutants and weather variables.
• Monte Carlo simulations of future pollution under customizable traffic trends (e.g., linear growth, historical patterns, green mobility).
• Visualization tools for diagnostics and station comparisons.
• Organized access to structured datasets at multiple spatial and temporal resolutions.

---

Installation

You can install the package locally by cloning the repository and running:

pip install .

Make sure you are using a Python environment with the following packages installed:

• pandas
• numpy
• seaborn
• matplotlib
• statsmodels

---

Modules Overview

summary() Function

The summary() function provides a flexible interface to run OLS regressions of pollutants on traffic data and meteorological conditions for selected monitoring stations in Ticino, Switzerland. It supports textual summaries and graphical diagnostics, as well as comparisons across stations and pollutants.

Parameters

Parameter	Type	Default	Description
station	str or list	"all"	Station name (e.g., "Airolo") or list of names. "all" includes all available stations.
show	str	"ols"	Output type: "ols", "graphic", or "all".
pollutants	str or list	"all"	Target pollutants. "all" includes: ["no", "no2", "nox", "pm10"].
comparison	bool	True	Whether to show summary tables and comparisons.

Example

from summary import summary
summary(station=["Airolo", "Bioggio"], show="all", pollutants=["no2", "pm10"])

---

monte_verita_simulation() Function

Simulates future pollutant concentrations under different traffic evolution scenarios using Monte Carlo methods.

Parameters

Parameter	Type	Default	Description
df	DataFrame	Required	Input with date, traffic, and pollutant columns.
variable	str	Required	Pollutant to simulate (e.g., "no2").
trend_type	str	"historical"	One of: "historical", "historical_modified", "linear", "exponential", "manual".
growth_rate	float	0.02	Used for exponential trend (e.g., 0.007974 = 10%/year monthly).
n_periods	int	48	Number of simulation periods.
n_simulations	int	1000	Number of Monte Carlo draws.
meteo_csv_path	str	Optional	CSV with monthly average temperature, wind, rain.

Example

from traffico_ticino.simulation import monte_verita_simulation

sim_df = monte_verita_simulation(
    df=df_bioggio_full,
    variable="no2",
    traffic_col="traffic",
    trend_type="exponential",
    growth_rate=0.007974,
    n_periods=60,
    meteo_csv_path="data/meteo_mensile/bioggio_meteo_mensile.csv",
    coef_json_path="data/coefficients/pooled.json"
)

---

load_dataset() Function

Helper function to load datasets from the data/ folder, including subdirectories and station-specific files.

Parameters

Parameter	Type	Description
nome_file	str	Base name of the file (without .csv).
sottocartella	str	Subfolder under data/.
stazione_singola	bool	If True, loads from data/stazioni singole/<sottocartella>.

Example

from traffico_ticino.loader import load_dataset

df = load_dataset("O3_media", "media")
df_station = load_dataset("BIASCA_traffic_only", "traffic_only", stazione_singola=True)

---

Project Directory Structure

traffico_ticino/
├── data/
│   ├── media/
│   ├── all/
│   └── stazioni singole/
│       ├── traffic_only/
│       ├── noO3/
│       └── ...
├── summary.py
├── simulation.py
├── loader.py
└── ...

---